A new AI translation system for headphones clones multiple voices simultaneously
Imagine going for dinner with a group of friends who switch in and out of different languages you don’t speak, but still being able to understand what they’re saying. This scenario is the inspiration for a new AI headphone system that translates the speech of multiple speakers simultaneously, in real time.
The system, called Spatial Speech Translation, tracks the direction and vocal characteristics of each speaker, helping the person wearing the headphones to identify who is saying what in a group setting.
“There are so many smart people across the world, and the language barrier prevents them from having the confidence to communicate,” says Shyam Gollakota, a professor at the University of Washington, who worked on the project. “My mom has such incredible ideas when she’s speaking in Telugu, but it’s so hard for her to communicate with people in the US when she visits from India. We think this kind of system could be transformative for people like her.”
While there are plenty of other live AI translation systems out there, such as the one running on Meta’s Ray-Ban smart glasses, they focus on a single speaker, not multiple people speaking at once, and deliver robotic-sounding automated translations. The new system is designed to work with existing, off-the shelf noise-canceling headphones that have microphones, plugged into a laptop powered by Apple’s M2 silicon chip, which can support neural networks. The same chip is also present in the Apple Vision Pro headset. The research was presented at the ACM CHI Conference on Human Factors in Computing Systems in Yokohama, Japan, this month.
Over the past few years, large language models have driven big improvements in speech translation. As a result, translation between languages for which lots of training data is available (such as the four languages used in this study) is close to perfect on apps like Google Translate or in ChatGPT. But it’s still not seamless and instant across many languages. That’s a goal a lot of companies are working toward, says Alina Karakanta, an assistant professor at Leiden University in the Netherlands, who studies computational linguistics and was not involved in the project. “I feel that this is a useful application. It can help people,” she says.
Spatial Speech Translation consists of two AI models, the first of which divides the space surrounding the person wearing the headphones into small regions and uses a neural network to search for potential speakers and pinpoint their direction.
The second model then translates the speakers’ words from French, German, or Spanish into English text using publicly available data sets. The same model extracts the unique characteristics and emotional tone of each speaker’s voice, such as the pitch and the amplitude, and applies those properties to the text, essentially creating a “cloned” voice. This means that when the translated version of a speaker’s words is relayed to the headphone wearer a few seconds later, it sounds as if it’s coming from the speaker’s direction and the voice sounds a lot like the speaker’s own, not a robotic-sounding computer.
Given that separating out human voices is hard enough for AI systems, being able to incorporate that ability into a real-time translation system, map the distance between the wearer and the speaker, and achieve decent latency on a real device is impressive, says Samuele Cornell, a postdoc researcher at Carnegie Mellon University’s Language Technologies Institute, who did not work on the project.
“Real-time speech-to-speech translation is incredibly hard,” he says. “Their results are very good in the limited testing settings. But for a real product, one would need much more training data—possibly with noise and real-world recordings from the headset, rather than purely relying on synthetic data.”
Gollakota’s team is now focusing on reducing the amount of time it takes for the AI translation to kick in after a speaker says something, which will accommodate more natural-sounding conversations between people speaking different languages. “We want to really get down that latency significantly to less than a second, so that you can still have the conversational vibe,” Gollakota says.
This remains a major challenge, because the speed at which an AI system can translate one language into another depends on the languages’ structure. Of the three languages Spatial Speech Translation was trained on, the system was quickest to translate French into English, followed by Spanish and then German—reflecting how German, unlike the other languages, places a sentence’s verbs and much of its meaning at the end and not at the beginning, says Claudio Fantinuoli, a researcher at the Johannes Gutenberg University of Mainz in Germany, who did not work on the project.
Reducing the latency could make the translations less accurate, he warns: “The longer you wait [before translating], the more context you have, and the better the translation will be. It’s a balancing act.”
